Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 85334
    Prev
    Next
    INDEX
    Explanations

    references to ideas and notable contributions in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    technical terms or ideas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    StoryboardSegue
    -0.79
     linkovi
    -0.70
     lenker
    -0.70
     مشين
    -0.66
     disambiguazione
    -0.66
     ProtoMessage
    -0.63
     resourceCulture
    -0.62
     الرياضيه
    -0.61
     OFDb
    -0.61
     שוליים
    -0.60
    POSITIVE LOGITS
    UPAC
    0.34
     munici
    0.33
    Finally
    0.33
    ついに
    0.32
    ukkig
    0.32
     finally
    0.31
    Modifica
    0.31
     empre
    0.31
     enviado
    0.31
    untung
    0.31
    Activations Density 0.027%

    No Known Activations